Floating-point multiplication

[Alex Holkner]

I'm experiencing discrepencies in floating point multiplication between the
Python interpreter and embedded python.  Running Python 2.2 under Win2k,
compiling with VC++.  The interpreter gives:

  >>> print sys.version
  '2.2 (#28, Jan  4 2002, 19:01:19) [MSC 32 bit (Intel)]'
  >>> print 0.785398163397 * 4.0
  3.14159265359
  >>> print math.pi
  3.14159265359

Which is expected.  Under my embeddeded python, however:

  print sys.version
    '2.2 (#28, Jan  4 2002, 19:01:19) [MSC 32 bit (Intel)]'
  print 0.785398163397 * 4.0
    3.14159274101
  print math.pi
    3.14159274101

Which is quite wrong, and is upsetting 'import random', which requires
math.pi to be within a certain tolerance.

I expect that this problem is related to the runtime C libraries in use,
however both the interpreter and embedded python22.lib/python22.dll were
compiled from the source.  Building Debug vs. Release does not make a
difference.

Thanks,
Alex.